BMI and proximity to a highway: Little evidence for an association

摘要

Background There is increasing interest in obesogenic chemicals that may increase BMI in exposed individuals. Particulate air pollution appears to have obesogenic potential and ultrafine particles (UFP, ＜100nm) are elevated next to highways. Aims We sought to use proximity to a major highway as a surrogate for exposure to UFP and test associations with BMI for participants in the Community Assessment of Freeway Exposure and Health (CAFEH) study. Methods This analysis is based on a combined random and convenience sample of 260 adults ＞40 years old living in two of the study areas for CAFEH. Multi-variate regression consisted of examining the association between proximity to highway and Body Mass Index (BMI) as a continuous variable. Model-building involved consideration of many variables, including demographics, environmental exposures, illnesses and medications, physical activity and diet. Univatiate regression was conducted to determine which variables were significant with the outcome BMI. Variables that had a p-value of .05 or greater were excluded from multi-variate regression analysis. Adjusted linear regression model building was performed using a backward, forward and stepwise selection approach with a p-value of 0.15 as the entry and exit criteria. Categorical distance from the highway was included in all models. Results The Beta Estimate for proximity to highway was 0.085 and the R square value was 0.0004. Vegetable Intake, physical activity and education level were found to be significant predictors that had an expected downward effect on BMI score and a reasonable contribution to the total model variability while being born in the US showed a significant increase in BMI. Conclusions We found little of an effect of proximity to highway on BMI. However, proximity is a limited surrogate for UFP exposure and additional analysis with assigned exposures is warranted.